ADAOI 113 DEFENCE RESEARCH ESTABL'SHMENT OTTAWA ONTARIO) " F le/8THE EFFECTIVENESS OF VARIOUS LOCAL SHIELDING ARRANGEMETS ER[ETTfC(u)SEP 79 F A JOHNSON, J BRISSON

UNCLASSIFIEO DRFO-TN-79-20

llll

oxr

RESEARCH AND DEVELOPMENT BRANCH

DEPARTMENT Of NATIONAL DEFENCE

CANADA

DEFENCE RESEARCH ESTABLI HIMENT OTTAWATEHNCA 1); "- /- TIC

TECHICALNOTE(NO./ 79-20 I,-CT

rB

JT HE EFFECTIVENESS OF YAR IQUS LOCAL_$HIELDINGk6RRANGEMENTS SRECTED NEAR TAfE IARGET OF THE

DAEO fPEUTRON GENERATOR IN J EDUCING THE.§IOLOGICALHAZ~ARD AT SVECIFIC LOCATIONS ON THE DRIE SITE,

tiv

t- .A. Johnson m#id .1i 1i isson

Nucleart Effects SectionProtective Sciences Division

1y. . _ -

Accesion ForUNCLASSIFIED

iTIS o

ABSTRACT

The effectiveness of various forms of shielding near the target ofthe neutron generator in reducing the dose rate at specific locations on theDREO site has been investigated. It would appear possible to satisfy AECBrequirements so far as reducing the dose rate along the perimeter fencebordering the railway right-of-way is concerned, but substantial additionalshielding would be required to reduce the dose rate significantly at the near-est occupied building, Building 14. In the latter case, to take advantage ofthe maximum operating hours permitted by the licence, AECB requirements couldbe met only by construction of a massive shield around the target, otherwisethe cumulative dose inside Building 14 could be monitored and operationsterminated when the total dose reached the permitted yearly limit for non-atomic workers.-

RESUME

Una etude a it6 faite sur l'efficacitf des diverses formes deblindage situfes pros de la cible du gfnfrateur de neutrons et servant Arfduire le dosage dan des endroits particuliers sur l'empl#cament du CRDO.Il semble possible de r~pondre aux exigences de la CCEA en ce qui a traitA la reduction du dosage le long du pfrimbtre de al cl8ture longeant le chtminde fer; cependant, pour rfduire de fagon appreciable le dosage I l'6dificevoisin, soit le batiment 14, it faudrait ajouter une quantitf considgralbe deblindage. Dane le pr6sent cas afin de tirer parti, au maximum des heures d'exploitation autorisfes par le permis, il nous faudrait, pour r6pondre auxattentes de la CCEA, construire un blindage massif autour de la cible, sinonil faudra contr8ler le dosage cumulatif A l'int6rieur de l'6difice 14 etl 'exploitation devra 8tre interrompue lorsque le dosage total aura atteint lalimite annuelle permise pour lee employfs qui no sont pas attaches aux travauxatomiques.

iii

UNCLASSIFIED

a ANK -NOT ?JE

UNCLASSIFIED

TABLE OF CONTENTS

1. INTRODUCTION....................................... .. .. .... .. . . . 1

2. RADIATION SURVEY TECHNIQUES .. .. .............. 3

3.* INITIAL REFERENCE SURVEY 3

3.1 Measurements .. .. ...... . . . . . . . .* 3

3.2 Derived Dose Equivalent of the Output of the

Generator. .............. .... 8

4. EFFECTS OF LOCAL SHIELDING NEAR THE TARGET 10

4.1 Shielding Configurations Employed . . . . . . . 10

4.2 General Effects of the Various ShieldingConfigurations .. .. ............ 10

4.3 Dose-Rate Reduction Along Railway Fence . . . . 12

4.4 Measurements at Building 14 . . .. .. .. . .12

5. THE EFFECTS OF SNOW-COVERED TERRAIN ON DOSE RATES MEASUREDAT OPEN-FIELD LOCATIONS. .. ............. 15

6.* RESPONSE CHARACTERISTICS OF THE EBERLINE PNR-4 REMCOUNTrER. .. ................. .... 16

7. SUMMARY. .. ......................... 18

8. ACKNOWLEDGEMENTS .. .. ................... 19

9.* REFERENCES . . . .. .. .. .. .. .. .. .. .. .. .. 20

V

UNCLASSIFIED

UNCLASS IFIED

1. INTRODUCTION

The DREO 150-keV Texas Nuclear Neutron Generator is situatedin an open-field location where it is shielded from occupied buildings on thesite by an L-shaped concrete-block wall and where the surrounding area is en-closed by a protective perimeter fence. On the basis of a previous radiationsurvey of the DREO site (1), for the case of a 14-HeV neutron output of 10 10n/s,the Atomic Energy Control Board expressed concern that the dose rate at theentrance of the nearest occupied building, Building 14, was of the order ofthe maximum rate to which non-atomic workers should be exposed at any time.The Board was also concerned that the dose rate at the perimeter fence alongthe railway right-of-way, which was in direct view of the target of the neutrongenerator, was much in excess of this maximum.

In its subsequent biennial review of the operating licence of thegenerator the Board imposed a number of specific conditions to which renewalof the licence was subject. Thus, for a 14-MeV neutron output of not morethan 3-4 x 109 n/s, suitable additional shielding, acceptable to the Board'sDirectorate of Licensing, was to be installed in the vicinity of the target inorder to reduce the dose-equivalent rate at Building 14. In other cases,where outputs of up to 101 n/s would be required, a shadow shield was also tobe erected between the target and the railway tracks, and a complete radiationsurvey was to be carried out to assess the effectiveness of the shield beforefurther operation at such output levels would be sanctioned. In addition, itwas requested that the effect, if any, of this shadow shield on the radiationlevel at the entrance to Building 14 be determined.

The present report describes the results of the radiation surveywhich was carried out in compliance with the above-mentioned directives. Inorder to assess the effectiveness of any shield which would be erected, adetailed neutron survey was first carried out over a measured grid within theperimeter fence and at certain locations elsewhere before any additionalshielding was put in place near the target. Then the effect of a number ofdifferent shielding configurations on the dose rate at various grid points wasdetermined including, in particular, those along the fence bordering the rail-way, and measurements were also made at the entrance to Building 14.

Details of the layout of the field installation, together with thelocations of the grid points at which survey measurements were made, are shownin Figure 1.

UNCLASSIFIED

2 UNCLASSIFIED

RAILWA R4.H1 0 WA1

PVIMAR"? To fENA H " LP t 4 4. 44t

0* 10 aA?!SVALS

TARG T

- V .AIN S#4LC4M! G WALL

4t Oc 4

P I RmAAOM 1 aNt I

Pigur I

i I ,tails of" the layout o'f the Neutron Genemr : l'ild insqtallation,

e;:ving/ locatione of" the grid points at which eu w' mea,emente

UNCLASSIFIED

UNCLASSIFIED 3

2. RADIATION SURVEY TECHNIQUES

The radiation survey was carried out using an Eberline portable,

battery operatedNeutron Rem Counter, Model PNR-4, the output of which wasrecorded from the phone connector by means of a scaler (Ortec Model 772).Since this survey was carried out mainly under winter conditions, over snow-

covered terrain, the scaler and its associated power-suppl.y bin were mountedon a toboggan, together with a Honda Model EG1500 gasoline-powered generator tosupply ac power. Figure 2 shows this mobile detection system in position at oneof the grid markers. For most of those measurements which were carried outduring cold weather the Ram counter, which consisted of a 22.86-cm (9-inch)diameter cadmium-loaded polyethylene sphere with a BFs tube in the center andan attached electronics module, was enclosed in a 2.54-cm (1-inch) thick in-sulating foam styrene cover which contained an electric light bulb of a suitablesize, depending on external ambient conditions, to maintain the temperature ofthe assembly above O°C. Because of small variations in speed of the engine ofthe generator, the frequency of the voltage output was not sufficiently con-stant to permit accurate electronic timing of counting intervals, and a stop-

watch was therefore employed for this purpose. In the few cases were mainspower was available and could be used in place of the generator, a Model 719Ortec Timer was used to time the counting period.

Calibration of the Rem counter was carried out using a standard Pu-Besource, and the output of the neutron generator was monitored by means of afission chamber whose calibration factor was established by suitable techniques.

This monitor was sufficiently close (40 cm at 1620) to the generator targetthat neutrons scattered from any shielding which would be erected in the vicinitywould not be expected to make a significant contribution to the total numberdetected, and thus determination of the target output would not be adverselyaffected.

3. INITIAL REFERENCE SURVEY

3.1 Measurements

Dose-rate measurements were made during the winter (January-February1978) at nearly all of the grid points shown in Figure 1, and also at the gateof the perimeter fence and at the entrance to Building 14. Measurements atsome locations near the target and at Building 14 were repeated later (April-May) after the snow had disappeared, and some additional measurements were alsomade at this time. The unshielded target of the neutron generator, as employedfor these measurements is shown in Figure 3A.

On the basis that the output of the T(dn)"He reaction is essentiallyIsotropic (within ".llZ) in the laboratory system (2), the results of all thesemeasurements can be most clearly and graphically sunuarized by means of a plotof the response of the Ram counter at a particular grid point versus the inversesquare of the distance of that grid point from the target. Such a plot is shownin Figure 4, where soma of the points are labelled with the particular gridlocations at which the measurements were made. In this figure two scales areprovided for the ordinate axis. The left-hand scale denotes the counts recorded

UNCLASSIFIED

* -~~-*~-..- ---- - MV

'4 UNCLASS ~ FlVP

I"

2' 'p'4 N

I.

.4

UNCLASSI I** 1EV

-F

4 S34

V\C~~~~ LAS .YIY1

UNCLASSI FLED

by the PNR-4 Rem counter normalized with respect to the fission chamber countsobtained during the same counting period. The right-hand scale represents thedose-equivalent rate corresponding to the left-hand scale (conversion factor0.149) and is based on the measured net response (100 cpm E 5.93 - 0.19 mrem/hr.)of the PNR-4 to a Pu-Be source of known output and on the calibration factor ofthe fission chamber (2.51 x l0 - 10'0 n/sec into 4n). Techniques by whichthese factors were determined will be described in a future report.

Apait from some measurements which were made at locations whose direct

view of the target was obstructed by the main L-shaped shielding wall near thetarget, such as those at E6 and F6, the majority of the points plotted inFigure 4 appear to fall into two relatively distinct groups such that the pointsin each group are distributed about, or conform closely to, an identifiablyseparate trend line. The two suggested trend lines, drawn through the origin,

have been labelled I and II. Thus for each of the two groups of points thedose rate varies approximately inversely as the square of the distance from thetarget, but a different apparent source strength is indicated for the two cases,as evidenced by the 'v 24% difference in the slopes of the lines I and 11. Thisdifference in the dose-distance dependence between the two groups of poinL isdue to a.number of factors, such as the auxilliary cement-block shielding nearthe target, the main shielding wall, the uneven terrain (particularly at backangles south of the target), and the (small) angular dependence of the outputfrom the target. While a distinct demarkation between the regions where themeasured dose rates follow one or other of the apparent trend lines is not tobe expected, an appraisal of the dose rates at certain grid points in relationto the locations of these points with respect to the target can suggest ageneral basis for the difference in the trends which were observed. The follow-ing observations can be made:

i) The measurements made at those grid points directly ahead ofthe target, C5, D5, E15 and F5. all lie close to the lower trendline II. In this case the shielding wall, while not obstructinga direct view of the target, acts to remove essentially half ofthose neutrons which would normally be expected to scatter fromthe ground into the detector, as well as a significant fractionof those neutrons which foim the skyshine component. In additionvery few neutrons reflected from the wall or from the auxillaryshielding near the target can be expected to reach the detectorat any of these positions. V

ii) The measurements at grid points G4 and G3, at % 900 to thetarget, also conform to the trend of line II. Since the outputfrom the target at this angle is expected to be no more than 6%

less than that at 00, and also because the neutrons detected atthese points would include those reflected from the auxillaryshielding near the target and from the wall, as well as theusual ground-scattered and skyshine components, it must beconcluded that the low dose-rates at these points is mainly dueto a reduced output from the target at this angle. A reductionin the neutron flux at 90" could occur because the neutrons areemitted in a direction essentially parallel to the plane of thetarget and can be absorbed or scattered in the target backing,in the cooling water, or In the target cap itself.

UNCLASS IF IED

UNCLASSIFIED 7

1 4 k. I

4~~~ H4 /21N

MKI

404

t4 /

00

Pi0r Sunn twdt4

Repoe ov th X- o one tvn-os,4 h ifiIi

twou/ th nee q i f(i o h ~ro~rgi

Ivint j~vm teuihs'e:tre.4w u vaie! itp

Wrmsit ho h ws5euisf.:a nfu~ 1,ytea0tktrend~~~~~ 3fe av~e k'1. 7: pf* 1,n. 4 r*o

mda 1"n of the PY-4ist ",'u'tes Y4 :ando of rspeo vey ao~tmoe~rvd~ them in'reFoftz ?z4. fte~:r~X'g

~~it y~~ ~a tsie.~ t~e. Te CIAiSSit'IFdistno

8 UNCLASSIFIED

iii) The measurement made at grid point 114 lies below trendline IT by a significant amount. This point representstrh_ largest backward angle (-130) at which any of themeasurements were made and, as such, would be the pointat which any effects of the non-isotropic output fromthe target would be most likely to be observed. Inaddition, the direct line-of-sight from the target to thispoint passes very close to the large oil-filled tank con-taining the isolation transformer and high-voltage powersupply associated with the neutron generator in BuildingTI, and hence neutron absorption or scattering by thetank could reduce the air- or ground-scattered fluxincident on the detector at this point, and absorptionby the building itself could reduce the direct flux.

iv) Those dose-rate measurements which appear to follow trendline I were made at points withii. the forward quadrant,in a direction away from the shielding wall. At thesepoints the major portion of the incident neutron flux,composed of the direct and the air- and ground-scatteredcomponents, would not be expected to be significantlyaffected by the presence of the wall. In addition, whilethese points were not equally accessible to neutronsreflected from the main wall, they all had an uaobstructedview of the auxilliary cement4lock shield between thetarget and the wall. Thi' shield was only about one metreaway from the target and would be a significant sourceof scattered neutrons.

Thus, in summary, the neutron flux described by trend line I, which ischaracteristic of a region in the forward quadrant away from the wall, is composedof direct and air- and ground-scattered components, together with a componentreflected mainly from auxilliary shielding near the target. The reduced fluxdescribed by trend line II, on the other hand, is characteristic of two differentregions where different factors contribute independently to produce apparentlysimilar intensity distributions. Thus in the direction around 0 the main shield-ing wall reduces the ground-scattered and skyshine components, and neutrons scat-fered from shielding near the target are absorbed by this wall, with the resultthat the flux in this direction is smaller than that described by trend line Ibut contains a larger proportion of the direct flux. In the other region around900 the flux is composed of the usual direct and scattered components character- Kistic of trend line I, but the intensity is smaller because the output from thetarget in this direction is reduced by absorption and scattering in the targetitself. At backward angles, greater than about 1200, effects of the non-isotropyof the output and other factors combine to cause the flux to become smaller thanthat suggested by trend line II.

3.2 Derived Dose Equivalent of the Output of the Generator

From the slopes of the two lines I and II drawn in Figure 4, dose-equiva-lent factors of 60.4 x l0- and 48.3 x 10-' mrem/n/cm2 respectively can be derivedas applicable to the output of the neutron generator in the regions characterized

UNCLASS IFIED

UNCLASSIFIED 9

by the two trend lines. The values derived for these factors depend in parti-cular on calibration of the Rem counter with respect to the calculated dose-equivalent factor (1%41.1 x 10-6 mrem/n/cm 2) of a standard Pu-Be source (averageEn 4.2 - 4.3 MeV) of known output. For 14-MeV neutrons a dose-equivalent factorof 57.8 x 106 mrem/n/cm 2 has been recommended (3), which corresponds closelywith that obtained from the slope of trend line I. However detailed comparisonof these factors is not meaningful for two reasons: the assumed flux of 1010 n/sfrom the target is augmented by an unknown amount by scattered neutrons whichare also incident on the detector, and the Eberline PNR-4 Rem counter does notprovide the appropriate relative dose-equivalent response for the differentneutron energies characteristic of the direct and scattered components. Thus,on the basis of the response curve published in the technical manual whichaccompanies the Rem counter (4), it is apparent that the measured responseapproximates the desired dose-equivalent curve only in the range of about 400keV to 3-4 MeV, being too high below this range and too low above. At 10 MeVthe response curve appears to be about 35% too low and falling rapidly; thusat 14 MsV the discrepancy could be expected to be even greater. An energyresponse of this nature appears to be typical of those neutron survey instru-ments which are available commercially (5).

The previous survey (1) which was carried out employed a differenttype of Rem counter, SNOOPY, which had similar response-characteristicdeficiencies to those exhibited by the PNR-4, and it is of interest to comparethe dose-equivalent rates which were measured at some common locations in thetwo surveys. Such a comparison is given in the following table:

LOCATION DOSE-RATE EQUIVALENT mrem/hour

Summer Conditions Winter Conditions

SNOOPY PNR-4 PNR-4

At gate of perimeter fence 0.3-0.5 0.36 ± 0.02

Railway fence at position ofminimum distance to target 1.5-1.7 2.03 ± 0.07

Outside entrance to Bldg. 14 0.27 ± 0.01 0.20 ± 0.02

Detector facing target 0.2-0.3

Detector facing up 0.1-0.2

UNCLASSIFIED

10 UNCLASS IFI ED

Thus the dose rates measured by the two instruments were similar, but precisequantitative comparison is not warranted since the dose rates measured withSNOOPY were obtained by visually estimating the average deflection of thefluctuating needle of the count-rate meter, whereas the output of the PNR-4was recorded by means of a scaler. It is apparent, however, that the measure-ments made with the PNR-4 at the entrance to Building 14 at different timesof year suggest that there is a significant reduction in the dose rate underwinter conditions when the surrounding area is covered with snow. This pointwill be discussed later. The errors quoted for the PNR-4 measurements in theabove table are statistical, being based on the number of counts obtained, andare not estimates of the accuracy of the dose rates themselves.

4. EFFECTS OF LOCAL SHIELDING NEAR THE TARGET

4.1 Shielding Configurations Employed

A number of different local shielding arrangements near the targetwere investigated in order to determine the most effective means of reducingthe dose rate at the specific locations identified by the AECB.

Three shielding configurations were employed as indicated in Figure 3B:

Si A 4 x 4 array of cement blocks, each 19 cm x 14 cm x 40 cm,located between the target and the railway tracks at a distanceof 30.5 cm from the centre of the target.

S2 A similar 4 x 5 array of cement blocks at a distance of 30.7 cmfrom the target, with three 18-cm-thick boxes of paraffin onthe side of the array away from the target and these surmountedby two cement blocks. This configuration takes advantage ofthe fact that compound shielding with heavier elements locatedtowards the incident neutron beam has been shown to constitutethe preferred elemental distribution (6).

S3 The same arrangement as S2, with the addition of a cement block43 cm square and 31 cm thick, with a centrally located hole 30.5cm In diameter, which was placed symmetrically about the target,with two of the standard cement blocks on top. In this case theshielding above the target subtended a solid angle of approxi-

mately 0.7-0.8 n steradians.

4.2 General Effects of the %arious Shielding Configurations

The dose-rate measurements which were made with these differentshielding arrangements are all plotted in Figure 5 versus the inverse squareof the distance of the particular grid point from the target. Both the norm-alized PNR-4 counts obtained at each of the grid points and the correspondingdose equivalents are indicated on the axis of ordinates. The straight lines

UNCLASS IFIED

UNCLASSIFIED 1

J.Adinfo 8/uogOI N3d unfoH 83d PI3M*W

0~ 0

too- 43

1~~4.11IL0

00

IL jt- c-0-j

toa' toC. 0r

-01 mc~j-0

O~O43

ot

cn

SLNflO: 8iO4.NOW 0,01 ki 3d SlNflO3 10- Nd

UNCLASSIFIED

12 UNCLASSIFIED

which have been drawn in Figure 5 are identical with those drawn in Figure 4

to represent the two trends of the dependence of dose rate with distancewhich were evident in the unshielded case.

Those points plotted in Figure 5 which appear to follow the trendssuggested by the lines which have been drawn represent measurements made at gridpoints whose direct view of the target was not obstructed by the various shield-ing structures erected near the target. Titus, on the basis that the apparenttotal neutron output from the target, as determined by the fission chamber,was not affected by scattered neutrons, it appears that the dose rate measuredat such grid points was not significantly influenced by the added shielding.The apparent anomaly presented by the measurements at grid point E4, wherein

shielding arrangements S2 and S3 slightly reduced the dose at this pointcompared with that measured in the unshielded case while arrangement Si had nosuch effect, can be accounted for by the fact that the 4 x 5 array of cementblocks in configurations S2 and S3 extended about 8 cm further forward of thetarget than did the 4 x 4 array in the Sl configuration. Thus, for shieldingconfigurations S2 and S3, the line joining the target to grid point E4 (and byextension to C3) approximately defines the transition between that regionsouth of the target which was shadowed from the target and the area forward of

the target which was not.

In the case of those grid points, such as G3, G4, F4, whose directview of the target was completely blocked by the added shielding, the degreeto which the dose rate was reduced was a function of the particular shieldingconfiguration, as indicated by the references to certain specific grid pointsin Figure 5. In general, the S2 configuration was markedly superior to SI,while S3 produced only a marginal improvement over S2.

4.3 Dose-Rate Reduction Along Railway Fence

As stated previously, one of the major points of concern to theAECB was the relatively high dose rate along the perimeter fence bordering therailway right-of-way which was produced by the unshielded target. Figure6 is a linear plot of the unshielded and shielded dose rates along this fence,as measured at grid points Al to Hl. These measurements are not contained inthe data plotted in Figure 4 and Figure 5, since such a presentation would beobscure and would not facilitate a convenient or direct visual comparison ofthe pertinent points. As can be seen from Figure 6, the shielding introducedbetween the target and the fence reduces the maximum dose rate by about afactor of 3 to a level of about 0.7 mrem/hr. It would appear that any furthersignificant reduction of the dose rate along the fence could only be attainedby a considerable increase in the amount of shielding over that already inplace.

4.4 Measurements at Building 14

Building 14 is located about 130 m from the target of the neutrongenerator and is shielded from the direct neutron flux by the main L-shaped

concrete-block wall near the target; thus the neutron flux incident on the

UNCLASS IFIED

UNCLASSIFIED 1

lfldJifo s/u 010 1 3d 8n0H 83d WM3&I'

U, 9 W

0 x 0 4soo

0 0

CLl

z t

I~0 IIN~

Ix 0 4 0Iz

N N4-

S.ZfO Z l1NII~I~~ JNO ~~~

UNCLASIF4E

_______<9__c_____1

14 UNCLASSIFIED

building will consist mainly of skyshine and ground-scattered components. Asstated previously, the AECB was concerned about the dose rate at the entranceto this building, as well as that inside the building since, on the basis ofthe previous survey (1), an integrated yearly exposure considerably greaterthan 5 mrem would result if the neutron generator was operated, even at areduced output of 1/3 x 1010 n/s, for a period approaching the total numberof hours (20 hours per week for 50 weeks) permitted by the operating licence(A18/77). Thus the effect, adverse or otherwise, of shielding added nearthe target on the dose rate at Building 14 was of interest to the AECB.

The measurements made at Building 14 are summarized in the followingtable, where the dose rates are normalized to an output of 1010 n/s from thetarget:

DOSE-RATE EqUIVALENT mrem/hour

TARGET SHIELDING CONFIGURATION SU101ER WINTERCONDITIONS CONDITIONS

Unshielded towards railway 0.28 - 0.01

0.26 - 0.01

Av.0.27 - 0.01 0.20 ± 0.02

S2 shielding 0.25 - 0.01

S3 shielding 0.18 - 0.01

The measurements listed in the table as being made under "summerconditions" were actually carried out in April-May (1978) after the snowcover had disappeared and as the shielding structure near the target wasbeing dismantled in stages. In this case the measurements made with thetarget unshielded towards the railway (i.e. the added shielding removed com-pletely) can be compared with those made with SNOOPY under similar conditions(i.e. absence of snow cover) in the earlier survey (1) and which have beendiscussed previously. It is apparent from the dose-equivalent rates undersummer conditions quoted in the above table that the shielding configurationS2 had very little effect on the dose rate at Building 14 as compared withthe unshielded case. It was only when shielding was introduced above thetarget, as was the case for the S3 shielding configuration, that a significantreduction in the dose rate occurred, the obvious mechanism being a reductionin the skyshine component. It should be noted, however, that the dose rateobtained with the S3 configuration under sumer conditions was almost thesame as that measured under winter conditions without any additional shielding

UNCLASSIFIED

r UNCLASSIFIED

near the target. In the latter instance it would appear that the snow coveracts to reduce the ground-scattered component incident on Building 14. Furtherexamples of the effect of the snow cover will be discussed below.

5. THE EFFECTS OF SNOW-COVERED TERRAIN ON DOSE RATES MEASURED AT OPEN-FIFI LOCATIONS

Different dose rates were measured at Building 14 under summer andwinter conditions for the same total neutron output rate from the target, asdiscussed above, and this difference was attributed to the effects of snowcover in reducing the ground-scattered flux. In this case it was possible toobserve the difference in dose rate clearly because only the skyshine andground-scattered flux were incident on the building, the direct flux beingblocked by the L-shaped shielding wall near the target. While the more-intensedirect flux would be expected to mask minor changes in the air- or ground-scattered components, it is possible that small seasonal changes in dose ratewere observed in some measurements at grid locations which were in direct viewof the target, as can be ascertained by inspection of Figure 4 (e.g. locationsE5, F4, G4, H4). More significant seasonal diferences, however, were observedfor points such as E6 and F6, which were shielded from the direct neutron fluxby the main wall and which were much closer to the target than had been thecase for Building 14. The dose rates measured at these two locations undervarious conditions, for a neutron output of 1010 n/s, are summarized in thefollowing table:

DOSE-RATE EQUIVALENT mrem/hour

TARGET SHIELDING E6 F6CONFIGURATION

Summer Winter Summer Winter

Unshlelded towards railway 2.24 ± 0.04 1.73 + 0.09 3.01 ± 0.06 2.28 ± 0.10

S2 2.04 ± 0.03 2.95 ± 0.06

S3 1.36 ± 0.03

1.35 ± 0.02

UNCLASSIFIED

16 UNCLASSIFIED

Thus, for the case of no added shielding near the target, the dose

rates at the two points are each about 23-24% smaller under winter conditionsthan in the summer, and these winter dose rates are also smaller than thosemeasured in the summer with the S2 shielding in place. The above results are

thus similar to those obtained at Building 14, but are statistically moresignificant because of the larger neutron flux available at the E6 and F6

locations.

6. RESPONSE CHARACTERISTICS OF THE EBERLINE PNR-4 REM COUNTER

As stated previously, the calibration of the Rem counter was basedon its response to a standard Pu-Be source of known output, this response beingrelated to the calculated dose-equivalent factor for the source. The responseof the Rem counter, however, approximates the desired dose-equivalent curveonly over a relatively small energy range (4), from 400 keV to 3-4 MeV, andthus, as pointed out, dose-rate measurements at locations where the direct14-4MV flux is incident on the counter would be expected to be in error by a

considerable amount.

A meaningful calibration of the response of the detector at 14-NeV,b3ased on the known output from the target as determined by the fission chamber

and the recommended (3) dose-equivalent factor applicable to 14-MeV neutrons,was not feasible at any of the unshielded locations at which measurements weremade because of the unknown contribution of the skyshine and ground-scattered

components. It was possible, however, to compare the dose rates indicated bythe PNR-4 Rem counter at a number of locations with those determined at thesame time by means of an NE-213 scintillation detector (7). In the lattercase the neutron spectrum above 0.5 HeY was determined by proton-recoil spectro-scopy and integrated against the ANSI (3) fluence-to-biological-dose conversionfactors. Comparisons were made both at unshielded locations uhere the 14-MeVflux from the target would be predominant, and also at locations where a directview of the target was blocked by the main shielding wall and where thae air-and ground-scattered flux would consequently be expected to make the majorcontribution to the measured dose rate.

The results of these measurements, normalized to a neutron outputof 1010 n/s, are summarized in the following table; also included are valuesfor the average neutron energy as determined in each case by the recoil protonspectrometer, as well as the expected dose rates for a pure 14-MeV flux calcu-lated on the basis of the recommended (3) dose equivalent factor of 57.8 10-6

mrem/n/cm2 for 14-MeV neutrons:

UNCLASS I FIED

Dis tance D O"-RJtlt Fuivlt trro/hour A\,.rage Rat i7

Location Target t,) %IaR- IT coi to d \ -

K3 _10.70 14.8 1 0. 1. c2

F4-F* 18.18 53.1 ± 0.$ . %0.1 .S 1. $

Fe, 24.8 . 2.95 ± 0,0e I l.' -_ ,.80

KOt 30 .37 :.!4 1 L1.0.1 1..4.,

* Halfway between grid p'oints F4 atid F, at About 3:o from target.

t S2 shielding in place, for all others target unshield*d towar-ds railway.

Thus at the first three locations litted in the table, all of whichhad an unobstructed view ot the target, the dose rates derived from the recoilspectrometer measurements are from 481 to WN' higher than those indicated bythe PNR-4 and reflect, in part. the inadequate dose-equivalent response of theRen counter to high-energy neutrons, as discussed previously. 1te fact thatthe recoil-proton measurements gt\ dosoe rate which are also constdorablyhigher than those calculated on the basis ot a 14-Z*V neutron output of 101%

n/s from the target can be attributed mainly to the contribution of scatteredneutrons. This conclusion ti supportod by the fact that the greatest difterene(57%) betw*n the calculated dose rates and those measurod by prtiAon recoiloccurs for the F4-F5 location. This particular location is the only one otthe throe which was within sight ot the auxtIlltary shielding wall near thetarget, and this wall wuld be expectod to be t significant soutrce of scatteredneutrons, as already pointed out. The loer ae\rage noutron energy. 11.5) MeV.which was measured at this location, as compared with that masu red at Fv orF5, is also consistent with this expectation. Thus the scattoret' -neutroncomponent makes up an important fraction of the total number of neutrons whtchwre detected by the recoil spectrometor at this location. The fact that, otthe PN-4 measurements, the one made at this same location deviates least kt%,from the calculated dose rate can be accounted for on the same basis. In thisinstance the greater sensitivity of the 'NR--4 to the lower-eonrgy flux ofscattered neutrons partially compnsates for its relative insensitivity to thehighor-energy direct component.

VNCLASSI FlED

18 UNCLASSIFIED

At locations F6 and E6, which are behind the main shielding wall andwhere the average incident neutron energy is less than 4 MeV, the dose ratesmeasured by the Rem counter are higher than those based on the recoil protonspectrum. At such locations it might be expected that the PNR-4 measurementswould be more reliable than those made at unshielded locations, since calibrationof the Rem counter was based on the known output of a standard Pu-Be sourcewhose average neutron energy of - 4.2 MeV is not greatly different from thatto be found at the shielded locations where these measurements were made. Thehigher dose rate indicated by the Rem counter as compared with that obtained bythe recoil proton technique can be attributed both to the enhanced response ofthe PNR-4 below 400 keV, assuming the measured spectrum is weighted relativelymore in this region than is the Pu-Be spectrum, and to the 500-keV detectionthreshold of the recoil proton spectrometer.

7. SUMMARY

A field survey has been carried out to assess the effectiveness ofvarious forms of shielding around the target of the neutron generator in re-ducing the dose rate at selected locations on the DREO site. Measurementsat specific locations of interest to the AECB showed that varying degrees ofadditional protection could be attained, depending on location. Thus, alongthe fence bordering the railway right-of-way, the most effective shieldingarrangement which was employed, S3, reduced the dose rate by only a factor of3 from that produced by the unshielded target. It was apparent that anyfurther significant reduction in dose rate would require the installation ofconsiderable additional shielding. At locations which were already shieldedby the main L-shaped wall near the target, such as Building 14, it was possibleto make a marginally significant reduction ( 50%) in the dose rate only whenshielding was introduced above the target. At such locations the beneficialeffects of snow cover in reducing the dose rate, through a reduction of theground-scattered component, were also observed; thus at Building 14 the winterdose rate without additional shielding near the target approximated thatobserved in summer with the maximum S3 shielding in place.

The results of this survey lead to the conclusion that the mosteffective shadow shield which was employed would probably satisfy AECB require-ments so far as the dose-rate along the railway fence was concerned, the AECBbeing prepared to accept the measured dose rate as being as low as reasonablyachievable (ALARA), considering the low occupancy factor. On the other hand,it can also be concluded that the marginal reduction in dose rate which wasproduced at Building 14 by means of shielding added above the target was in-dicative that even a more substantial shield of the same nature would not betotally acceptable under the AECB guidelines to permit continued operation ofthe generator for the full number of hours specified in the licence. Thus itcan be surmised that AECB requirements as regards the yearly dose at Building 14

UNCLASSIFIED-- - - -- - '

UNCLASSIFIED 19

could only be met either by construction of a massive concrete shield whichsurround the target practically entirely, or else by monitoring factorsrelated to the neutron output in order to obtain a measure of the cumulativedose inside Building 14, and being prepared to cease operations for the yearonce the total had reached 5 mrem.

The practical utility of the PNR-4 Rem Counter for carrying out ameaningful quantitative survey can be questioned as a result of experiencewhich has been gained in the use of the instrument in the field. While theinstrument itself can apparently be calibrated very precisely and consistentlywith Pu-Be neutrons, its inadequate relative response to a different neutronspectrum suggests that dose-rate measurements made with the instrument in afast-neutron flux of arbitrary spectral composition can be relied upon to nomore than a factor of two.

8. ACKNOWLEDGEMENTS

The authors wish to thank Mr. J.R.W. Boyle for his assistance inlaying out the survey grid, Mr. L.P. Rushton for assisting in some of themeasurements, and Mr. F.G. Strain for operating the neutron generator duringcalibration of the fission detector.

UNCLASSIFIED I

20 UNCLASSIFIED

9. REFERENCES

1. A Radiation Survey of Selected Portions of the DREO site to Determinethe Biological Hazard Presented by Operation of the DREO NeutronGenerator in an Oven-Field Location.F.A. Johnson and F.G. StrainDREO Technical Note No. 77-12.

2. Neutron Production Cross Sections and Energies for the ReactionsT(pn)3 He, D(d,n)sHe, and T(d,n)VHe.H. Liskien and A. PaulsenNuclear Data Tables 11 (1973) 569.

3. American National StandardNeutron and Gamma-Ray Flux-to-Dose-Rate FactorsANSI/ANS-6.1.1 - 1977 (N666).

4. Technical ManualPortable Neutron Rem Counter Model PNR-4Eberline Instrument CorporationSanta Fe, New Mexico.

5. The Energy Response of Some Commercial Beta/Gamma and Neutron SurveyInstruments.I.M.G. Thompson, A. Lavender, R.G. Shipton and J. GoodwinProc. Symp. Advances in Physical and Biological Radiation Detectors.Vienna (1970)Paper IAEA-SM-143/38.

6. Shielding Measurements for 3 MeV and 15 MeV NeutronsJ.J. Broerse and F.J. Van WervenHealth Phys. 12 (1966) 83.

7. H.A. Robitaille and B.E. Hoffarth,Defence Research Establishment Ottawa, private communication.

UNCLASS IFIED

UNCLASS IFIED%&Culity Classifiction

DOCUMENT CONTROL DATA R & 0St " , amy~tall ..... t~l 1-41 , it n,.t.4,!.u.i.. )99, 44 I1 ,,, , , hen 1hit ovitl l,:a nnti

I ORICONAIING ACIIVII!, 2. DOCUMENT SECURITYV CLASSIFICA! UN

Defence Research Establishment Ottawa UnclassifiedNational Defence Headquarters It,' GROUPflttAian flpin, CAnnAA IClA InnA

3 DOCUMENT TITLE

The Effectiveness of Various Local Shielding Arrangements Erected near the

Tage DOUMN Dte NeOtro NOerto OF PAdci gE he Niooia OFaRarS

atSepmeri 1979 ion co 17 n St

Septembr 17720 1 1

llAO0 92

8b CONTRACT NO 9lb. OTH4ER DOCUMENT NO(IS) (Any othei mbets that mav heassignedi t his cint-ument I

10 DISTRIBUTION STATEMENT

Distribution unlimited

11 SUPPLEMENTARY NOTES 12. SPONSORING ACTIVITY

13. ABSTRACT(U

The effectiveness of various forms of shielding near the target ofthe neutron generator in reducing the dose rate at specific locations onthe DREO site has been Investigated. It would appear possible tosatisfy AECB requirements so far as reducing the dose rate along theperimeter fence bordering the railway right-of-way is concerned, butsubstantial additional shielding would be required to reduce the doserate significantly at the nearest occupied building, Building 14. In thelatter case, to take advantage of the maximum operating hours permittedby the licence, AECB requirements could be met only by construction of amassive shield around the target, otherwise the cumulative dose insideBuilding 14 could be monitored and operations terminated when the totaldose reached the permitted yearly limit for non-atomic workers.

031S

I77

Security Classiicatlion

KEY WORDS

Fast neutronsNeutron fluxNeutron scatteringNeutron shieldingRem countersDose-equivalent rateNeutron detectors

INSTRUCTIONS

I ORIGINATING ACTIVITY Enter thi, name and address of the 9b. OTHER DOCUMENT NUMBERiS), If the document has beenorgarrzar-on issuing the dlocument. assigned any other document numbers (either by the originator

or by the sponsor), also enter thts number(s).2,1 DOCUMENT SECURITY CLASSIFICATION Enter the overall

security Llassification of the document including special warning 10. DISTRIBUTION STATEMENT Enter any hmirations onterms whenever applicable. further dissemination of the document, other than tho e imposed

by security classification, using standard statements such as2t GROUP Enter security reclassification group number. The three

groups a,e defined in Appendix 'M'of the ORB Security Regulations. Ill "Ouatified requesters may obtain copies of thisdocument from their defence documentation center."

S DOCUMENT TITLE Enter the complete document title in allcapital letters Titles in all cases should be unclassified. If a (2) "Announcement and dissemination of this documentsufficiently descriptive title cannot he selected without classil- is not authorized without prior approval from,:ation. show title classification with the usual one-capital-letter originating activity "ahtbrev,atorn purrrntheses immedilrrrly followirng the title.

11. SUPPLEMENTARY NOTES Use for addit,onal explanatoryZ DESCRIPTIVE NOTES Enter the , rlegory of document. e.g. notes.

techncal report. technical note or t,-,hnica letter If appropri-ire, enter the type of document, e q ,nterim. progress, 12. SPONSORING ACTIVITY Enter the name of the departmentalsummary, annual or final Give the rclusve dates when a project office or laboratory sponsoring the resparch andspecific reporting period is covered development. Include address.

5. AUTHOR(Sf: Enter the namels) of author(s) as shown orr or 13. ABSTRACT Enter an abstract grving a brief and factualin the document. Enter last name, first name. middle initial. summary of the document, even though it may also appearIf military. show rank. The name of the principal author is an elsewhere in the body of the document itself It is highlyabsolute minimum requrrement. desirable that the abstract of classified documents be unclassi

fhed. Each paragraph of the abstract shall end wrth an6. DOCUMENT DATE Enter the date Imonth, year) of indication of the security classrfication of the informatron

Establrshment approval for publrcation of the document. in the paragraph (unless the document itself is unclassified)represented as (TS), (S), (C). (R), or (U).

7a. TOTAL NUMBER OF PAGES The total page count shouldfollow normal pagination procedures, r.e., enter the number The length of the abstract should be limited to 20 single-spacedof pages containing information. standard typewritten lines, 712 inches long.

7b. NUMBER OF REFERENCES: Enter the total number of 14. KEY WORDS: Key words are technrcally meaninful terms orreferences crted in the document, short phrases that characterize a document and cruld be helpful

rn cataloging the document. Key words should be selected soBa. PROJECT OR GRANT NUMBER' If appropriate, enter the that no security classrfrcatron is required. Identifers, such as

applrcable research and development prolect or grant number equipment model designation. trade name. mirtary project coderinder which the document vwas wrrtirn. name, geographic location, ma

, be used as key words but will

be followed by an indication ol technical context.Bb CONTRACT NUMBER If appropriate, enter the applicable

rrumbe, under which the eloirment was written.

9a. ORIGINATOR'S DOCUMENT NUMBER(S): Enter theofficial documr'nt number by whrch the document will beidentified and controlled by the orlrnatng activrty. Thrsnumber must be unique to this do ,ment.

"AD-A0BI 123 DEFENCE RESEARCH ESTABLISHMENT OTTAWA (ONTARIO) F/6 18/6

TIE EFFECTIVENESS OF VARIOUS LOCAL SHIELDING ARRANGEMENTS ERECT--ETC(U)

SEP 79 F A OHNSON, J BRISSONWdCLASSjIFIjED DREO-TN-79-20 R

I D

SUPPLEMENTA

INFORMATION

DEFEWE RESEARCH ESTABLISfMT OTTAWA

DREO TEHNICAL NOTE NO. 79-20

IM EFFECTIVENESS OF VARIOUS LOCAL

SHIELDING ARRANGBI4ENT ERECTED NEARTHE TARGET OF ThE DREO N JIRON GENERATOR

IN REDUCING THE BIOLOGICAL HAZARD AT

SPECIFIC LOCATIONS ON ThE DREO SITE

by

F.A. JCHNSON AND J.R. BRISSON

ERRATA SHEET

* Figure 3A and figure 3B photographs should by interchanged but

* the captions should remain as they are.

Page 6, line 6 should read "the fission chamber (2.51 x 105 cpm

1010 n/sec into 4). Techniques by which ... "

Page 17, table should be F6t and not just F6.

,34